TTOpt: A Maximum Volume Quantized Tensor Train-based Optimization and its Application to Reinforcement Learning
Machine Learning
2022-09-29 v2 Neural and Evolutionary Computing
Optimization and Control
Abstract
We present a novel procedure for optimization based on the combination of efficient quantized tensor train representation and a generalized maximum matrix volume principle. We demonstrate the applicability of the new Tensor Train Optimizer (TTOpt) method for various tasks, ranging from minimization of multidimensional functions to reinforcement learning. Our algorithm compares favorably to popular evolutionary-based methods and outperforms them by the number of function evaluations or execution time, often by a significant margin.
Cite
@article{arxiv.2205.00293,
title = {TTOpt: A Maximum Volume Quantized Tensor Train-based Optimization and its Application to Reinforcement Learning},
author = {Konstantin Sozykin and Andrei Chertkov and Roman Schutski and Anh-Huy Phan and Andrzej Cichocki and Ivan Oseledets},
journal= {arXiv preprint arXiv:2205.00293},
year = {2022}
}
Comments
26 pages, 8 figures, accepted to Thirty-sixth Conference on Neural Information Processing Systems (NeurIPS 2022). Pre camera-ready version